Evaluating Chain-of-Thought reasoning in large language models for thyroid ultrasound interpretation: a dual-information approach.

Front Artif Intell · 2026 · PMC13050882 · PMID 41948697

Europe PMC DOI

hedged sentences

0.0010

closest p · 0.0× alpha

0.0530

boldest claim

The sentences

highly significantp < 0.001actually significant

Authenticity of CoT responses by reasoning-capable LLMs For reasoning authenticity, the Kruskal-Wallis H test demonstrated highly significant differences in physician ratings among the four models ( p < 0.001).

also in 132,142 other papers

did not reach statistical significancep = 0.053so close (0.05 < p ≤ 0.1)

Although ChatGPT-O3 showed numerically higher accuracy than Grok-3, this difference did not reach statistical significance after BH correction ( p = 0.053).

also in 111,027 other papers

Quoted from the open-access full text in Europe PMC under the licence the publisher applied. The sentence is reproduced exactly as published; the emphasis is ours.